Best practices

Building effective agents is about creating systems that think, act, and improve over time. This section provides strategic and technical guidance to help you design agents that are accurate, resilient, and production-ready. Whether you're designing your first agent or refining a live deployment, follow these best practices to guide design decisions, prevent common errors, and ensure long-term maintainability.

The following sections walk you through key dimensions of agent development:

• Agent vs workflows: Understand when to use an agent, a workflow, or a hybrid approach.
• Best practices for building agents: Learn how to scope, prompt, and evaluate effectively.
• Choosing the best model for your agent: Test and select the best model for your use case.
• Best practices for publishing and deploying agents: Validate agents for production and track quality over time.
• Best practices for context engineering: Ensure your agents retrieve accurate and relevant knowledge.
• Best practices for DeepRAG and Batch Transform: JIT vs. index-based strategies